Geochemical and Grain-Sized Implications for Provenance Variations of the Central Yellow Sea Muddy Area Since the Middle Holocene

AI Lina, HAN Zongzhu, WU Xiao, 2), SAITO Yoshiki, and WANG Houjie, 2), *

Geochemical and Grain-Sized Implications for Provenance Variations of the Central Yellow Sea Muddy Area Since the Middle Holocene

AI Lina1), HAN Zongzhu1), WU Xiao1), 2), SAITO Yoshiki3), 4), and WANG Houjie1), 2), *

1),,,266100,2),,266237,3),,690-8504,4),,305-8567,

Based on high-resolution analysis to a 280-cm long sediment core obtained from the muddy area in the central Yellow Sea, we examined the provenance of muddy sediments and discussed the changing marine sedimentary environment since the middle Holocene. The results indicated that fine-grained sediments in the muddy area were mainly derived from the Huanghe (Yellow River) and Changjiang (Yangtze River) with considerable stepwise variations during the past 6.6kyr. The Yellow Sea Warm Current was initiated at 6kyr when the sea level was high together with the enhanced East Asian Winter Monsoon. These in combination established the framework of shelf circulation in the Yellow Sea that began to trap the river-derived fine-grained sediments. From 4.9kyr to 2.8kyr, both the Kushiro Current and East Asian Monsoon were significantly weakened, reducing the delivery of Changjiang sediments to the muddy area. As a result, the sediments were mainly originated from the Huanghe. From 2.8kyr to 1.5kyr the continuously weakened East Asian Winter Monsoon and enhanced Yellow Sea Warm Current entrapped more fine-grain sediments. Whereas the enhanced East Asian Winter Monsoon and the human caused increase in sediment load of the Huanghe since 1.5kyr, and direct delivery of Huanghe sediments to the Yellow Sea during 1128–1855 AD might dominated the sedimentation in the study area. The stepwise variations of the sediment provenance and composition of the Central Yellow Sea muddy sediments are of importance to understanding the formation of muddy deposit in the central Yellow Sea and the associated variations of marine environment since the middle Holocene.

sediment provenance; rare earth elements; grain size; central Yellow Sea mud; middle Holocene

1 Introduction

The Yellow Sea (YS), as a typical epicontinental sea, is located at the margin of the north Western Pacific (Qin., 1989). It is a receiving basin for a large amount of terrestrial sediment delivered by the large rivers in China and relatively smaller rivers in western Korean Peninsula (Yang., 2003; Liu., 2009) with several large muddy patches distributed in the central and western Yellow Sea (Shi., 2003; Wang., 2010; Bian., 2013; Li., 2014). Among those, the central Yellow Sea mud (CYSM) patch is evidently separated from coastal zone, as characterized by fine-grained sediments accumulated since the middle Holocene when the sea level became high and stable (Milliman., 1986; Alexander., 1991; Shi., 2003; Yang and Youn, 2007). The fine-grained sediments preserved in the muddy patch act as important archives for deciphering the evolution of marine environment and indicating the dispersal system of terrestrial sediments. Therefore, the CYSM has been extensively investigated during the past several decades with variable topics ofstratigraphic sequence, reconstruction of sedimentary history and the changing terrestrial sediment supply (Wang., 2014; Lim., 2016; Zou., 2016; Hu., 2018).

The source-to-sink processes in CYSM have been extensively investigated, however, debates on the provenance still remain. Previously published documents indicated that the sediments of the CYSM were primarily contributed by the Huanghe (Yellow River) that delivered a huge amount of terrestrial sediment to the marginal sea (., Milliman., 1986; Qin., 1989; Alexander., 1991; Zhao., 2001; Liu., 2004; Yang and Liu, 2007). In contrast, other studies proposed that the sediment of CYSM was a mixture from multi-sources including the inputs from Huanghe, Changjiang (Yangtze River) and small rivers in the western Korean Peninsula, and even from the abandoned Huanghe Delta where sedi- ment resuspension was active in a highly energetic environment (Li., 2014; Wang., 2014; Lu., 2016). It also has been confirmed that the provenance of sediments in CYSM was complex and varied temporally and spatially with changing river sediment delivery, channel migration, shelf circulation system and climatic oscillations during the Holocene (Yang and Youn, 2007; Li., 2014; Wang., 2014; Lim., 2016). However, the sediment provenance of the CYSM, especially since the middle Holocenewhen the sea level reached its highs, has remained poorly understood so far, as well as the source-to-sink processes. Therefore, the identification of sediment provenance has become critical to understanding the evolution of muddy deposits in the central Yellow Sea since the middle Holocene. More substantial evidences from the sediment records (., multiple indices from mineralogy, geochemistry and sedimentology) are therefore necessary to reliably discriminate the sediment sources in the CYSM deposition (Yang., 2003; Li., 2014; Lim., 2016).

Geochemical approaches attached importance to bulk sediments have been widely used as potentially excellent tools to discriminate marine sediment sources and decipher sedimentary environment (Cullers., 1987; McLennan., 1989; Holser, 1997; Vital., 1999; Lan., 2009). As elemental concentrations of sediments are mostly constrained by the grain-size compositions of sediments (Yang., 2007; Jung., 2016), for example, geochemical compositions tend to vary with grain size in the Changjiang and Huanghe sediments, it should be prudent to use the geochemical proxies for interpreting sediment provenance. Selection of relevant grain-sized fraction in the source areas is thus a necessary precondition of the reasonable comparison between the CYSM sediments and the potential sources. The fine-grained sediments (<2μm), which is the most active component for long-distance delivery from the source region to the shelf sea, have shown prominently superiority in provenance studies (Wan., 2010; Li., 2014; Xu., 2014; Zhao., 2018). But there are still considerable uncertainties for distinguishing sediments in the potential mixed sources areas such as CYSM, using clay mineral assemblage and content because of the complex hydrodynamic processes and similarity of clay mineral compositions in the Huang- he and Changjiang riverine sediments (Yang., 2003). Rare Earth Elements (REEs)are characterized by strongly partitioned and highly conservative behaviors in the pro- cesses of rocks weathering, surface erosion and transport along hydrological pathway (Taylor and McLennan, 1985), and have been accepted as a reliable tool for determining depositional processes and sediment provenance, as well for understanding the paleo-environmental evolution in more detail. In addition, geochemical characteristics of sediments controlled by clay mineral are more important than those controlled by heavy minerals (Yang., 2002, 2003; Xu., 2009). As a result, REE compositions of fine-grained sediments (<2μm) are the sensitive proxy indicator to discriminate sediment provenance, and this method has rarely been used to the CYSM.

Here we present comprehensive data sets including accelerator mass spectrometry (AMS)14C-dating ages, high- resolution grain size data, REEs characteristics of fine- grained sediments (<2μm) for core HS1 from the CYSM zone. The main purposes of this study include: 1) to characterize grain size and geochemical compositional variations of the core sediments, 2) to establish provenance proxy indicators to distinguish fine-grained sediments, and then constrain the sediment sources of the CYSM and their temporal changes, and 3) to elucidate the relationship between provenance variations and variable river inputs, channel shifts, shelf circulation system evolution, climate changes, anthropogenic activities, and fluctuations of East Asian Monsoon, with emphasis on the role of ocean currents in the sediment transport process.

2 Regional Setting

The Yellow Sea, bordered by mainland of China and Korean Peninsula, is characterized by asymmetric morphological feature with broad and flat sea floor deepened progressively southeastward with an average water depth of 55m and a maximum depth of 100m (Qin., 1989; Yang., 2003). Generally, the Yellow Sea Trough (YST) confined by 80-m isobaths is elongated in a north-to- south direction in the central YS (Qin., 1989). The YS has experienced a rapid Holocene marine transgression, as the sea levels reached close to its present level at about 6.0–7.0calkyrB.P. (Liu., 1999; Li., 2009), and mud deposits have been formed since the early Holo- cene (Liu., 2007). For example, the CYSM in the central South Yellow Sea (SYS), where the water depth locally exceeds 70m, is about 61×103km2in area with a relatively low sedimentation rate (1.4mmyr−1), and 228.8×106tons of sediments deposit annually in this muddy area (Qiao., 2017).

The ocean currents in the YS are characterized by eddy circulation and upwelling induced by the north-westward Yellow Sea Warm Current (YSWC) along the western side of YST, and the southward Yellow Sea Coastal Current (YSCC) and Korean Coastal Current (KCC) along the Chinese and Korean coasts (Zang., 2003; Fig.1). The YSWC is a shelf branch of the Kuroshio Current (KC), and transports warm and saline water toward the YS in winter (Wu., 2009). The variation of the YSWC is dominated by the wind stress curl in winter and Kuroshio Current (KC). Wind stress curls could force variations of the YSWC surface axis, whereas the stronger KC could drive the YSWC extending northward against the strong northerly winds in winter (Song., 2009; Wang., 2012). The YSCC is driven by the surface wind over the YS, which is dominated by the East Asian Winter Monsoon (EAWM), flowing southward along the 40–50m isobaths (Zang., 2003). In winter, sediments resuspended from the delta and coastal regions by storms are transported southeastward along the wind-enhanced YSCC (Yuan., 2008). In the central YS, cyclonic baroclinic circulation of the Yellow Sea Cold Water Mass (YSCWM) reinforces the shelf circulation in summer (Zhang., 2008). The ocean currents, located at the west and east sides of YS, present significant impacts on the transport and fate of fine-grained sediments in the YS (Huang., 2010; Wang., 2012a; Zhou., 2015; Yuan., 2018).

Fig.1 Schematic map showing the locations of core HS1 in the present study and other typical cores for reference (ZY2, Wang et al., 2011; B03, Hu et al., 2014; Z1, Pi et al., 2016; EZ06-6, Lim et al., 2016; H07, Leng et al., 2017; YSC-1, Li et al., 2014; B-3GC, Jian et al., 2000) in the south Yellow Sea and adjacent areas. The shelf circulation (arrow) as well as the muddy deposition areas (shaded polygon) are modified after Yuan et al., 2008 and Li, 2005. SDCC, Shandong Coastal Current; YSCC, Yellow Sea Coastal Current; CDW, Changjiang Diluted Water; ECSCC, East China Sea Coastal Current; TWWC, Taiwan Warm Current; KC, Kuroshio Current; TWC, Tsushima Warm Current; YSWC, Yellow Sea Warm Current; KCC, Korea Coastal Current.

The YS receives a large supply of terrestrial sediments and freshwater from the adjacent continental rivers. The Huanghe is located on the North China Craton, draining about 0.8×106km2with sediment discharge of 1.1×109tyr−1(Ren and Shi, 1986; Alexander., 1991; Satio., 2001). Most of the Huanghe-delivered sediments were trapped in the proximal subaqueous delta (Wright., 1988), with only 1%–15% (Alexander., 1991; Martin., 1993) of the total discharge being exported out of the Bohai Sea (BS) and into the YS. In addition the Huang- he has changed its lower channel both regionally and locally. There have been 11 channel shifts documented, at least twice when Huanghe flowed in the south of the Shandong peninsula during the Holocene (Saito., 2001; Xue., 2004). The Changjiang, with a drainage basin area of 1.8×106km2, annually delivers 928 km3of freshwater and 4.68×108t of sediments to the sea (Chen., 2001; Liu., 2006). Most of the Changjiang sediments discharged to the sea were trapped in its estuary and partly escaped and moved southeastward along the East China Sea inner shelf (Liu., 2006). However, Gao. (1997) proposed that the YSWC transported about 1×106tyr−1of sediments from the Changjiang into the YS, consistent with the previously reported Changjiang-derived clay minerals found on the seafloor in eastern south YS and central north YS (Lin., 1992). In addition, many small rivers from China and Korea Peninsula annually delivered a total sediment load of less than 50×106tyr−1into the YS (Lim., 2007). These rivers have diminutive contribution to sediment input due to their relatively low sediment loads (Yang., 2003).

3 Materials and Methods

3.1 Core Sampling

Sediment core HS1 was collected using a stainless steel gravity core sampler deployed by theFig.1). Core HS1 was 280cm in length and no sediments were lost or distorted in the upper layers during the processes of sampling and storage. The core was composed of gray to dark gray clayed silt. Core HS1 was sampled every 1cm for grain size analysis. A total of 56 samples were selected with 5-cm interval to extract the fine grained fraction (<2μm) for geochemical analysis.

3.2 Chronological Analysis

Mixed species of benthonic foraminifera from four samples were picked for accelerator mass spectrometry (AMS)14C dating at Beta Analytic Company, USA. The measured radiocarbon ages were corrected for the regional marine reservoir effect (Δ=−138±(−38)yr) and calibratedto calendar ages before 1950 AD (calB.P.), using the online program CALIB 7.1, with an updated calibration curve Marine13 (Reimer., 2013). Calendar ages are given with two standard deviation (2δ) uncertainty (Table 1).

3.3 Grain Size and Geochemical Analysis

The sediment samples were pretreated with 10% H2O2to decompose organic matters and 0.5molL−1HCl to remove carbonates. The grain size analysis was conducted using a Mastersizer 2000 laser particle-size analyzer at the Key Laboratory of Submarine Geosciences and Prospecting Techniques, College of Marine Geoscience, Ocean University of China, with a measurement range from 0.02 to 2000μm and analytical precision better than 3%. The grain size parameters were calculated following the method of McManus (McManus, 1988).

Table 1 AMS 14C ages and calendar ages of core HS1

Geochemical analysis was performed on the fine-grained sediment fraction (<2μm) of 56 samples in core HS1. All samples were pretreated to remove organic matters with 15% H2O2for 24h. The fine-grained fraction (<2μm) of each sample was extracted by conventional Stokes law and then concentrated by centrifuging. Size-separated sedi- ment samples were oven-dried at 60℃, and then powdered and homogenized with an agate mortar. The powdered sediments (50mg) were digested with a mixture of HF-HNO3-HClO4for 48h in a tightly closed Teflon beaker on a hot plate at less than 180℃. Then each sample was reacted with 1mL HNO3to remove the residual HF, and digested with a mixture of 5mL HNO3(50%) and 1mL Rh (500ppb) for 24h in a tightly closed Teflon vessel in an oven at 150℃. Each solution was analyzed for concentrations of REEs using an inductively coupled plasma-mass spectrometer (ICP-MS, Thermo X series) at the Key Labo-ratory of Marine Hydrocarbon Resources and Environmental Geology, Qingdao Institute of Marine Geology. The analytical accuracy was monitored by repeated analysis of China Stream Sediment Reference Materials (GSD9) along with the analysis of blanks. Relative deviations be- tween measured and known values were generally less than 5%.

4 Results

4.1 Depositional Rates and Sedimentary Sequence

According to the AMS14C dating data (Table 1 and Fig.2a) and paleo-environmental studies in the study area, sediments in core HS1 preserved the sedimentation information in a shelf environment during high sea level period after 7kyrB.P. The average sedimentation rate is estima- ted to be 44.75cmkyr−1that is comparable with the data derived from the adjacent cores of ZY-2 (Hu., 2012b), YSC-1 (Li., 2014), EZ06-1 (Lim., 2016), Z1 (Pi., 2016), H07 (Leng., 2017).

Fig.2 (a) The AMS 14C ages of core HS1; (b) Profiles of sand/silt/clay percentage, mean grain size and sorting coefficient of core HS1.

The sediment in core HS1 is mainly composed of silt and clay with the average contents of 61.2% and 38.3%, respectively, and there is no or very little sand with the content less than 0.6% (Fig.2a). The profiles of mean grain size and sorting coefficient present significant fluctuations with value ranges of 4.1–6.4μm and 1.3–1.6, respectively. Three deposition units can be identified with evident differences in both grain size and sorting coefficient (Fig.2b). Unit 3 (280–190cm) presents an intensive fluctuation in grain size and sorting degree. Unit 2 (190– 100cm) shows a relatively stable sedimentary environment. The sediments in the lower part of Unit 2 are finer than the upper part. Unit 1 (100–0cm) is characterized by an upward coarsening tendency in the upper part (0–40cm) and a downward coarsening tendency (40–60cm). And the sediment sorting of Unit 1 still presents a similar trend to mean grain size. Moreover the grain size frequency distribution curves for sediments in this core show little change and display a uni-modal pattern (Fig.3).

Fig.3 Spectrum of grain size distribution of sediment samples at different depths.

4.2 Geochemical Characteristics of Fine-Grained Sediment (<2μm)

The distributions of total rare earth element concentrations (ΣREE), the ratio of light rare earth elements to heavy rare earth elements (LREE/HREE), δEuN, δCeUCC, (La/ Yb)UCC, (La/Sm)UCC, and (Gd/Yb)UCCthrough the core are shown in Fig.4. According to the REE parameters, the sediment of core HS1 can be divided into three units (Units 1–3). The ΣREE fluctuate frequently and range between 174.9 and 226.8μgg−1, with the higher values in Unit 1 and Unit 3 (202.3μgg−1and 212.6μgg−1in average, respectively) and the lowest values in Unit 2 (191.2μgg−1in average). The distribution of δEuN, (La/Yb)UCC, (Gd/ Yb)UCChave the similar trend to that of ΣREE, with the average values of 0.67, 1.15 and 1.26 respectively. In contrast, δCeUCCwith meanvalue of 0.96 is opposite to ΣREE in Unit 2, and relatively stable in Unit 1 and Unit 3. Average values of LREE/HREE and (La/Sm)UCCare 9.7 and 0.94, showing relatively stable through the core, whereas higher values are found at 20–35cm. Detailed characteristics of REE compositions for each unit, as well as the main potential provenances, are given in Table 2.

Fig.4 Distributions of REEs fractionation parameters in the core HS1: ΣREE, LREE/HREE, δEuN, δCeUCC, (La/Yb)UCC, (La/Sm)UCC, and (Gd/Yb)UCC.

Table 2 Summary of geochemical characteristics of the core HS1 and the main rivers in different provenances

The upper continental crust (UCC) normalized REE fractionation patterns display a weak convex trend in terms of atomic number, with a general middle REE enrichment in all three units, which are similar to that of the sediments derived from the Changjiang and Huanghe, differing largely from the sediments derived from the Korean rivers with relative LREE enrichment and HREE depletion. In detail, the UCC-normalized REE patterns for the sediments of Unit 2, which had lower concentration of REEs, are more similar to the Huanghe sediments, while those of other units were relatively closer to the Changjiang sediments (Fig.5).

Fig.5 UCC-normalized REE fractionation patterns of Core HS1 and main potential provenances.

5 Discussion

5.1 Sediment Provenance Discrimination Based on Geochemical Evidences

Accurate provenance identification is crucial for understanding sedimentation processes from sediment source to sink, sediment dispersal patterns and paleo-environmental changes in the YS (Yang., 2003; Yang and Youn, 2007; Jung., 2016; Greggio., 2017; Flood., 2018). Geochemical properties of sediments can be used as potential proxies for distinguishing the contributions of Chinese rivers from those of Korean rivers in the YS (Yang., 2002, 2003; Song., 2009; Xu., 2009). In particularly, REEs have been accepted as a reliable tracer for determining depositional processes and sediment provenance in marine environments, according to their coherently behaviors with each other during surface processes, their stability during sedimentation and dia- genetic processes (Taylor and McLennan, 1985; Liu., 2009; Dou., 2010; Xu., 2012; Lim., 2016).

The distribution pattern of the UCC-normalized REEs suggested that the sediments in the core HS1 are not from the single source. The sediments in Unit 2 showed an evi- dent similarity to the Huanghe-derived sediments, while the sediments in Units 1 and 3 were similar to Changjiang sediments, all obviously distinguished from Korean river sediments (Fig.5). The (La/Sm)UCC. (Gd/Yb)UCCbinary diagram (Fig.6) indicated that Unit 2 samples are overlapped with Huanghe sediments, suggesting the dominant contribution from the Huanghe River. In contrast, the sedi- ments in Units 1 and 3 presented signals scattering between the Changjiang and Huanghe sediments, possibly indicating the mixed products mainly from the two Chinese rivers. In addition, sediments transported by the Korean rivers differed largely from those in core HS1 in geochemical proxy, implying that the Korean rivers had little contribution to the mud deposits in the central Yellow Sea. This result was also corresponded to the previous speculations that Chinese river sediments were mostly settled on the Chinese coast, while Korean river sediments were confined to the coast along Korean Peninsula (Lim., 2014; Jung., 2016; Koo., 2018).

Fig.6 Discrimination plot of (La/Sm)UCCversus (Gd/Yb)UCCfor the fine grained sediments (<2μm) of core HS1. Valuesof the Changjiang (<2μm) (Qiao and Yang, 2007), Huang- he (<2μm) (Ta, 2014), Korean rivers (suspended sediments) (Yang et al., 2003).

It seemed to be very difficult to distinguish the Changjiang and Huanghe sediments by using binary diagram of two parameters alone as they had fairly similar and overlapped REE compositions (Yang., 2003). To identify the contributions of Changjiang and Huanghe to the sedi- ments in the core accurately, the Multidimensional Discriminant Function () was defined as the discrimination index for Changjiang and Huanghe sediments, respectively. Based on the principle of Euclidean distance of spatial vector, the geochemical parameters of each sample could be expressed as spatial vector in the multidimensional space, and theis the distance between two spatial vectors, samples and end-members, in fact. The lower thevalues, the more likely the sample derives from that end-member (Razjigaeva and Naumova, 1992; Yang., 2002).

whereMDFis the distance between the sampleand end-memberin-dimensional space,xis the value of parameterin sample,xis the value of parameterin end member,Sis the standard deviation of parameter.

In this case, the total six geochemical parameters (ΣREE, (La/Yb)UCC, (Gd/Yb)UCC, (La/Sm)UCC, δEuN, δCeUCC) had been used to calculate thefor two end members, the Changjiang and Huanghe, respectively (Fig.7). The results indicated that the Unit 2 corresponded lowervalues for the Huanghe end-member and higher values compared with Changjiang end-member, suggesting the Huanghe fluvial sediments were dominant in Unit 2. Ne- vertheless, Units 1 and 3 preserved thesignals suggesting to be from both the Changjiang and Huanghe sediments, which indicated mixed sources of sediments from the Changjiang and Huanghe. Our results provided a geochemical interpretation for the stepwise variations of fine-grained sediment provenance in the YS during the Holocene that was generally consistent with the previous understanding from clay mineralogy and geochemical evidences (Yang., 2003; Li., 2014; Zhou., 2015; Lim., 2016).

5.2 Implications to the Changing Paleo-Environment

The close linkage between paleo-environmental changes and variations of detrital sediment supply could be addressed through grain-size compositions, clay mineralogy, and geochemical characteristics in surface and core sediments from the YS and the adjacent seas. These information in combination could be used to trace sediment pro- venance, sediment transport and accumulation processes, and also to provide information on the weathering regime on the continents(Yang., 2003; Yang and Youn, 2007; Wang., 2010; Wang., 2014; Lim., 2016). Core HS1 from the CYSM preserved the fine- grained sediments deposited in this area since 6.61kyr, corresponding to high sea level period, and the core can be divided into three distinct units (Units 1–3) with different geochemical characteristics of fine-grained detrital sediments (Fig.7). Due to the highly complex nature of sediment source and dispersal pattern in the CYMS (Yang., 2003; Yang and Youn, 2007; Liu., 2009; Dong., 2011; Wang., 2014), many factors may have controlled detrital sedimentation in the YS, including the shelf circulation, changing terrestrial supply of sediment in context of past global change and even the channel shifts of Huanghe (Xue., 2004; Lim., 2007; Liu., 2010; Hu., 2014; Wang., 2014; Zhou., 2015; Lim., 2016; Hu., 2018). The stepwise variation of sediment provenance identified from core HS1 in CYSM provided important information to discuss how the paleo-environment changed since the middle Holocene.

Sensitive populations of sediments are believed to contain key information on both the sediment source and hy- drodynamics (., McCave., 1995; Boulay., 2003; Sun., 2003).The standard deviation method was selected to partition the sensitive populations from the grain-size distribution of sediments (Xiao., 2006; Liu., 2010), which allowed a convincing identification of grain-size intervals with the highest variability along a sedimentary sequence. The standard deviations could be calculated for 280 samples in core HS1 based on different grain size classes. The plot of standard deviation values. grain size classes showed several peaks, and each peak represented a population of grains with the highest variability through time (Figs.8 and 9). Two grain size populations for sediments in core HS1 were identified,., fine population (<8.5μm) and coarse population (>8.5μm) (Fig.8). As shown in Fig.9, both the sensitive population contents and grain size parameters of sediments in core HS1 varied significantly, which might represent the record of the dynamic environment changes (Boulay., 2003; Sun., 2003; Xiang., 2006; Liu., 2010).

Fig.7 Multi-dimensional Discriminant Function (MDF) of core HS1.

Fig.8 Decomposition of the grain size distributions based on standard deviation method.

Unit 3 (6.61–4.91kyr) corresponded to the sedimentation in the middle Holocene when the local sea-level reached its highest level and became stable (Liu., 2004). The shelf circulation was established in the YS around 6.0kyr (Liu., 1999; Li., 2009) when the YSWC began to intrude into the YS (Li., 2009), which played a critical role in the formation of the Holocene muddy depositions in YS. As a result, the volume of fine populations in Unit 3 of core HS1 correspondingly increased and then became relatively stable at depth of 250–190cm during 6.0–4.9kyr (Fig.9). At this stage (6.61–4.91kyr) the fine-grained sediments derived from both the Huanghe and Changjiang began to accumulate in the central YS to initiate the formation of muddy patch (Figs.7 and 9; Yang and Liu, 2007; Yang and Youn, 2007).

In Unit 2 (4.91–2.87kyr), the volume of the fine populations reached its highest value (67%) at depth of 190– 100cm, corresponding to the lower mean grain size (<5μm; Fig.9). The changes of grain-size compositions dur-ing this period were also recorded in the cores close to core HS1, such as ZY2 (Hu., 2012) and YSC-1 (Li., 2014) in the central YS (see Fig.1). Therefore, the paleo-environment seemed to experience an abrupt changeat 4.91kyr. The transport and accumulation of fine-grained sediments in the CYSM were highly modulated by the shelf circulation that was closely associated with the intensity of KC and EAWM, the former of which dominated the intensity of YSWC and the latter modulated the coastal current along the western Yellow Sea (YSCC; see Fig.1). We combined the datasets including Sea Surface Temperature (SST), intensity of KC and EAWM estimated based on the sediment cores from the CYSM (., ZY2 and Z1; see Fig.1) and the muddy patch to the southwest of Cheju Island (., B3, see Fig.1) from previous papers. The Holocene variations of East Asian Summer monsoon (EASM) and EAWM were represented by the changing oxygen isotopic compositions extracted from both the stalagmite of Dongge Cave (Wang., 2005) and Dunde ice core (Shi., 1994; Fig.10). These combined datasets together with the data of fine population in core HS1 indicated that both the KC and EAWM became weakened from 4.7kyr to 2.8kyr even though there were considerable fluctuations (Fig.10; Jian., 2000). As a result, the SST derived from the sediment cores of ZY2, Z1 and B3 presented a decreasing trend with fluctuations. The weakened KC and EAWM possibly decreased the intensity of shelf circulation (., YSWC and YSCC), which therefore weakened the northward transport of sedi- ments from the Changjiang. The weakened EAWM was coincident with warm period recorded in Dunde ice cores (Fig.10; Shi., 1994). As a result, the sediment pro- venance of CYSM at this stage was mostly dominated by the Huanghe (Figs.6 and 7).

Fig.9 Variations of fine and coarse populations, and mean grain sizes of sediment particles for core HS1.

However, the upper part (140–100cm) in Unit 2 presented upward coarsening tendency during 3.78–2.87kyr with the mean grain size of 4.8μm, evidently differing from the lower part (190–140cm, with the mean grain size of 4.5μm; Fig.9). This might be resulting from the human interventions on the Huanghe at 4kyr as the Great Yu initially accomplished the channelization of the Huang- he (Yuan., 1998; Wu and Ge, 2005; Chen., 2012), which effectively increased the capacity of sediment delivery to the sea compared with the significant deposition in the flooded fluvial plain before human interventions (Huang., 2012; Zhang., 2013). As a result, more coarse particles were perhaps delivered to the sea and accumulated in the CYSM, corresponding to the increase in grain size of sediments in the upper part of Unit 2 since 3.78kyr (Figs.9 and 10).

Fig.10 Variations of (a) content of fine populations in core HS1, compared with (b) the evolution of EAWM in core ZY2 (Hu et al., 2012), (c) the evolution of EASM in Dongge Cave (DA) stalagmite (Wang et al., 2005), (d) δ18O of Dunde ice core (Shi et al., 1994), (e) the evolution of KC in core B-3GC (Jian et al., 2000), (f) SST derived from core ZY2 (Wang et al., 2011), (g) SST derived from core Z1 (Pi et al., 2016), (h) SST derived from core B3 (Zhao et al., 2014).

In Unit 1 (<2.87kyr), the YSWC became powerful as indicated by the enhanced KC and increasing SST in se- diment cores from CYSM (Fig.10), which could favor the northward transport of sediments from the Changjiang to the CYSM (Wang., 2011; Li., 2014; Zhao., 2014; Lim., 2016). Therefore, the sediment provena- nce in Unit 1 were the mixed sediments from both Chang- jiang and Huanghe as inferred from the geochemical proxies (Figs.6 and 7). There was a cooling process since 2.9kyr (Fig.10; Shi., 1994). The volume of the fine po- pulation increased to 67% at 1.5kyr, and decreased subsequently together with the enhanced EAWM that would physically sort the surface sediment in a way of active resuspension (Figs.9 and 10; Liu., 2010; Hu., 2012; Pi., 2016). The increase of sediment grain size since 1.36kyr in Unit 1 coincided with the human caused increase of sediment loads from the Huanghe since 1.5kyr when human activities in the Loess Plateau became significantly enhanced (., Wu, 2016). Meanwhile, the lower Huanghe course was artificially shifted southward from the BS to the SYS in 1128 AD in order to prevent the nomadic intrusion and thus a large delta lobe formed in the coast of Jiangsu Province (Fig.1) until 1855 AD when the Huanghe reentered the BS (Xue., 2004). The proximity of CYSM to the Huanghe delta during 1128– 1855 AD favored a large amount of coarser sediments being delivered to and eventually accumulated in the CYSM (Liu., 2004, 2007; Wang., 2009), which also contributed much to the increase of sediment grain size in Unit 1 of Core HS1 since 1kyr (Figs.9 and 10).

6 Conclusions

Based on the stepwise variation of sediment grain size, the sediment core (HS1) could be divided into three units from bottom to top: Unit 3 (6.61–4.91kyr), Unit 2 (4.91– 2.87kyr), and Unit 1 (<2.87kyr). The results of REE com- positions and the calculation of multidimensional discriminant functions indicated that fine-grained terrige- nous sediments in the study area mainly originated from the Changjiang and Huanghe, whereas the Korean rivers has little contribution to the muddy deposits in the central Yellow Sea. In Unit 2 (the middle part), the sediment was dominated by the terrestrial supply from the Huanghe, whereas the Units 1 and 2 preserved the mixed sediments from both the Changjiang and Huanghe. The variation of fine-grained sediment population (<8.5μm) indicated that the Yellow Sea Warm Current was initiated at 6kyr when the sea level was high, together with the enhanced East Asian Winter Monsoon. All of these established the frame- work of shelf circulation in the Yellow Sea that began to trap the fine-grained sediments from both the Changjiang and Huanghe. From 4.9kyr to 2.8kyr (Unit 2) both the Kushiro Current and East Asian Winter Monsoon were significantly weakened, which decreased the delivery of Changjiang sediment to the muddy area in the central Yellow Sea. As a result, the sediments in Unit 2 were mainly transported from the Huanghe, as confirmed by the REE analysis. In Unit 1 (the most upper part, 2.8kyr to present) the REE analysis indicated the mixed sediment sources from both the Changjiang and Huanghe. From 2.8kyr to 1.5kyr the continuously weakened East Asian Winter Monsoon and enhanced Yellow Sea Warm Current entrapped more fine-grained sediments, whereas since 1.5kyr the East Asian Winter Monsoon enhanced, resulting in the upward coarsening tendency of sediments. Meanwhile, the anthropogenically-increased sediment load from the Huanghe since 1.5kyr and the direct delivery of Huanghe sediments to the Yellow Sea during 1128–1855 AD might impact the sedimentation at this stage. The stepwise variations of sediment provenance and composition in the CYSM addressed in this paper is of importance to understanding the formation of muddy deposit in the central Yellow Sea and the associated variations of marine environment since the middle Holocene.

Acknowledgements

We appreciate the kind support from the crew of

Alexander, C., DeMaster, D., and Nittrouer, C., 1991. Sediment accumulation in a modern epicontinental-shelf setting: The Yellow Sea., 98 (1): 51-72.

Bian, C. W., Jiang, W. S., and Greatbatch, R. J., 2013. An exploratory model study of sediment transport sources and deposits in the Bohai Sea, Yellow Sea, and East China Sea., 118: 5908-5923.

Boulay, S., Colin, C., Trentesaux, A., Pluquet, F., Bertaux, J., Blamart, D., Buehring, C., and Wang, P., 2003. Mineralogy and sedimentology of Pleistocene sediments on the South China Sea (ODP Site 1144)., 184: 1-21.

Chen, Y., Syvitski, J. P., Gao, S., Overeem, I., and Kettner, A., 2012. Socio-economic impacts on flooding: A 4000-year history of the Yellow River, China., 41: 682-698.

Chen, Z., Li, J., Shen, H., and Wang, Z., 2001. Yangtze River of China: Historical analysis of discharge variability and sediment flux., 41 (2-3): 77-91.

Cullers, R. L., Barrett, T., Carlson, R., and Robinson, B., 1987. REE and mineralogical changes in Holocene soil and stream sediment: A case study in the Wet Mountains, Colorado, USA., 63: 275-297.

Dong, L. X., Guan, W. B., Chen, Q., Li, X., Liu, X., and Zeng, X., 2011. Sediment transport in the Yellow Sea and East China Sea., 93 (3): 248-258.

Dou, Y. G., Yang, S. Y., Liu, Z. F., Clift, P., Yu, H., Berne, S., and Shi, X., 2010. Clay mineral evolution in the central Okinawa Trough since 28ka: Implications for sediment provenance and paleo-environmental change., 288 (1-4): 108-117.

Flood, R. P., Barr, I. D., Weltje, G. J., Roberson, S., Russell, S., Meneely, J., and Orford, J. D., 2018. Provenance and depositional variability of the thin mud facies in the lower Ganges- Brahmaputra Delta, West Bengal Sundarbans, India., 395: 198-218.

Gao, S., Park, Y. A., Zhao, Y. Y., and Qin, Y. S., 1996. Transport and resuspension of fine grained sediment over the sou- theastern Yellow Sea. In:. Seoul National University Press, Seoul, 83-98.

Greggio, N., Giambastiani, B. M. S., Campo, B., Dinelli, E., and Amorosi, E., 2017. Sediment composition, provenance, and Holocene paleoenvironmental evolution of the southern Po River coastal plain (Italy)., 53 (3): 914- 928.

Huang, C. C., Pang, J., Zha, X., Zhou, Y., Su, H., Wan, H., and Ge, B., 2012. Sedimentary records of extraordinary floods at the ending of the mid-Holocene climatic optimum along the upper Weihe River, China., 22 (6): 675-686.

Huang, D., Zhang, T., and Zhou, F., 2010. Sea-surface temperature fronts in the Yellow and East China Seas from TRMM microwave imager data., 57 (11-12): 1017-1024.

Holser,W. T., 1997. Evaluation of the application of rareearthelements to paleoceanography., 132(1-4): 309-323.

Hu, B., Li, J., Zhao, J. T., Yan, H., Zou, L., Bai, F., Xu, F., and Wei, G., 2018. Sr-Nd isotopic geochemistry of Holocene sedi-ments from the South Yellow Sea: Implications for provenance and monsoon variability., 479: 102-112.

Hu, B., Yang, Z., Qiao, S., Zhao, M., Fan, D., Wang, H., Bi, N., and Li, J., 2014. Holocene shifts in riverine fine-grained se- diment supply to the East China Sea distal mud in response to climate change., 24 (10): 1253-1268.

Hu, B., Yang, Z., Zhao, M., Saito, Y., Fan, D., and Wang, L., 2012. Grain size records reveal variability of the East Asian Winter Monsoon since the middle Holocene in the central Yellow Sea mud area, China., 55 (10): 1656-1668.

Jian, Z., Wang, P., Saito, Y., Wang, J., Pflaumann, U., Oba, T., and Chen, X., 2000. Holocene variability of the Kuroshio Current in the Okinawa Trough, northwestern Pacific Ocean., 184 (1): 305-319.

Jung, H., Lim, D., Jeong, D., Xu, Z., and Li. T., 2016. Discrimination of sediment provenance in the Yellow Sea: Secondary grain-size effect and REE proxy., 123 (1): 78-84.

Koo, H. J., Lee, Y. J., Kim, S., and Cho, H., 2018. Clay mineral distribution and provenance in surface sediments of central Yellow Sea mud., 22 (6): 989-1000.

Lan, X. H., Zhang, X., Zhao, G., and Zhang, Z., 2009. Distributions of rare earth elements in sediments from core NT1 of the South Yellow Sea and their provenance discrimination., 38 (2): 123-132 (in Chinese with English abstract).

Leng, C. X., Yuan, H. J., Hao, Y., and Zhao, G., 2017. Applications of logarithm ratio transformation to extraction of the sensitive grain size of East Asian Winter Monsoon by the method of factor analysis: A case study of core H07 from the central mud area of the South Yellow Sea., 37 (1): 151-161 (in Chinese with English abstract).

Li, G., Yang, Z., and Liu, Y., 2005.. Science Press, Beijing, 1-65.

Li, J., Hu, B., Wei, H., Zhao, J., Zou, L., Bai, F., Dou, Y., Wang, L., and Fang, X., 2014. Provenance variations in the Holocene deposits from the southern Yellow Sea: Clay mineralogy evidence., 90 (1): 41-51.

Li, T., Nan, Q., Jiang, B., Sun, R., Zhang, D., and Li, Q., 2009. Formation and evolution of the modern warm current system in the East China Sea and the Yellow Sea since the last degla- ciation., 27 (2): 237-249.

Lim, D. I., Choi, J. Y., Jung, H., Rho, K., and Ahn, K., 2007. Recent sediment accumulation and origin of shelf mud depo- sits in the Yellow and East China Seas.,73 (2): 145-159.

Lim, D., Xu, Z., Choi, J., Li, T., and Kim, S., 2016. Holocene changes in detrital sediment supply to the eastern part of the central Yellow Sea and their forcing mechanisms., 105 (1): 18-31.

Lin, C., 1992. A study on the source and the distribution of clay at the Yellow Sea., 1 (2): 41-51 (in Chinese with English abstract).

Liu, J., Li, A., Xu, K., Velozzi, D., Yang, Z., Milliman, J., and Demaster, D., 2006. Sedimentary features of the Yangtze Ri- ver-derived along-shelf clinoform deposit in the East China Sea., 26 (17-18): 2141-2156.

Liu, J., Li, S., Wang, S., Yang, Z., and Ge, Z., 1999. Sea level changes of the Yellow Sea and formation of the Yellow Sea Warm Current since the last deglaciation., 19 (1): 13-24 (in Chinese with English abstract).

Liu, J., Milliman, J., Gao, S., and Cheng, P., 2004. Holocene development of the Huanghe’s subaqueous delta, North Yellow Sea., 209 (14): 45-67.

Liu, J., Saito, Y., Kong, X., Wang, Y., and Zhao, L., 2009. Geochemical characteristics of sediment as indicators of post- glacial environmental changes off the Shandong Peninsula in the Yellow Sea., 29 (7): 846-855.

Liu, S., Shi, X., Liu, Y., Qiao, S., Yang, G., Fang, X., Wu, Y., Li, C., Li, X., Zhu, A., and Gao, J., 2010. Records of the East Asian Winter Monsoon from the mud area on the inner shelf of the East China Sea since the mid-Holocene., 55 (21): 2306-2314.

Liu, Z., Colin, C., Huang, W., Chen, Z., Trentesaux, A., and Chen, J., 2007. Clay minerals in surface sediments of the Pearl River drainage basin and their contribution to the South China Sea., 52 (8): 1101-1111.

Martin, J., Zhang, J., Shi, M., and Zhou, Q., 1993. Actual flux of the Huanghe (Yellow River) sediment to the Western Pacific Ocean., 31 (3): 243-254.

McLennan, S., 1989. Rare earth elements in sedimentary rocks: Influence of provenance and sedimentary processes., 21 (1): 169-200.

McCave, I. N., Manlghettl, B., and Beverldge, N. A. S., 1995. Circulation in the glacial North Atlantic inferred from grain- size measurements., 374 (6518): 149-152.

Milliman, J., Fan, L., Zhao, Y., Zheng, T., and Richard, L., 1986. Suspended matter regime in the Yellow Sea., 17 (3-4): 215-227.

Pi, Z., Li, T., and Nan, Q., 2016. Environmental changes since mid-Holocene revealed by core Z1 in the South Yellow Sea and their response to East Asian Monsoon., 32 (7): 1-10 (in Chinese with English abstract).

Qiao, S., and Yang, Z., 2007. Comparison of rare earth element compositions in different grain size fractions of sediments from the Yangtze and Huanghe., 27 (6): 9-16 (in Chinese with English abstract).

Qiao, S., Shi, X., Wang, G., Zhou, L., Hu, B., Hu, L., Yang, G., Liu, Y., Yao, Z., and Liu, S., 2017. Sediment accumulation and budget in the Bohai Sea, Yellow Sea and East China Sea., 390 (1): 270-281.

Qin, Y., Zhao, Y., and Chen, L., 1989.. China Ocean Press, Beijing, 256-285.

Razjigaeva, N., and Naumova, V., 1992. Trace element composition of detrital magnetite from coastal sediments of northwestern Japan Sea for provenance study., 62 (5): 802-809.

Reimer, P., Bard, E., Bayliss, A., Beck, J., Blackwell, P., Ramser, C., Buck, C., Cheng, H., Edwards, R., Friedrich, M., Grootes, P., Guilderson, T., Haflidason, H., Hajdas, I., Hatte, C., Heaton, T., Hoffmann, D., Hogg, A., Hughen, K., Kaiser, K., Kromer, B., Manning, S., Niu, M., Reimer, R., Richards, D., Scott, E., Southon, J., Staff, R., Turney, C., and Johannes, P., 2013. IntCal13 and Marine13 radiocarbon age calibration curves 0–50,000 years cal BP., 55 (4): 1869- 1887.

Ren, M., and Shi, Y., 1986. Sediment discharge of the Huanghe (China) and its effect on the sedimentation of the Bohai and the Yellow Sea., 6 (6): 785-810.

Saito, Y., Yang, Z., and Hori, K., 2001. The Huanghe (Yellow River) and Changjiang (Yangtze River) deltas: A review on their characteristics, evolution and sediment discharge during the Holocene., 41 (2-3): 219-231.

Shi, X., Shen, S., Yi, H., Chen, Z., and Meng, Y., 2003. Modern sedimentary environments and dynamic depositional systems in the southern Yellow Sea., 48 (z1): 1-7.

Shi, Y., Kong, Z., Wang, S., Tang, L., Wang, F., Yao, T., Zhao, X., Zhang, P., and Shi, S., 1994. The climatic fluctuation and important events of Holocene megathermal in China., 37 (3): 353-365.

Song, Y., and Choi, M., 2009. REE geochemistry of fine-grained sediments from major rivers around the Yellow Sea., 266 (3-4): 328-342.

Song, D., Bao, X., Wang, X., Xu, L., Lin, X., and Wu, D., 2009. The inter-annual variability of the Yellow Sea Warm Current surface axis and its influencing factors., 27 (3): 607-613.

Sun, Y., Gao, S., and Li, J., 2003. Preliminary analysis of grain- size populations with environmentally sensitive terrigenous components in marginal sea setting., 48 (2): 184-187.

Taylor, S., and McLennan, S., 1985.. Blackwell, Oxford, 632-633.

Ta, J., 2014. Geochemical records of clay-grain size sediments and their implications for the provenance and sedimentary. Master thesis. Ocean University of China.

Vital, H., Stattegger, K., and Garbe-Schonberg, C., 1999. Composition and trace element geochemistry of detrital clay and heavy mineral suites of the lowermost Amazon River: A pro- venance study., 69 (3): 563- 575.

Wan, S., Li, A., Clift, P., Wu, S., Xu, K., and Li, T., 2010. Increased contribution of terrigenous supply from Taiwan to the northern South China Sea since 3Ma., 278 (1-4): 115-121.

Wang, H., Bi, N., Saito, Y., Wang, Y., Sun, X., Zhang, J., and Yang, Z., 2010. Recent changes in sediment delivery by the Huanghe to the sea: Causes and environmental implications in its estuary., 391 (3-4): 302-313.

Wang, Y., Li, G., Zhang, W., and Dong, P., 2014. Sedimentary environment and formation mechanism of the mud deposit in the central South Yellow Sea during the past 40kyr., 347: 123-135.

Wang, F., and Liu, C., 2009. An N-shape thermal front in the western South Yellow Sea in winter., 27 (4): 898-906.

Wang, F., Liu, C., and Meng, Q., 2012. Effect of the Yellow Sea Warm Current fronts on the westward shift of the Yellow Sea warm tongue in winter., 45: 98- 107.

Wang, L., Yang, Z., Zhang, R., Fan, D., Zhao, M., and Hu, B., 2011. Sea surface temperature records of core ZY2 from the central mud area in the South Yellow Sea during last 6200 years and related effect of the Yellow Sea Warm Current., 56 (15): 1588-1595.

Wang, Y., Cheng, H., Edwards, R., He, Y., Kong, X., An, Z., Wu, J., Kelly, M., and Dykoski, C., 2005. The Holocene Asian Monsoon: Links to solar changes and North Atlantic Climate., 308 (5723): 854-857.

Wright, L., Wiseman, W., Bondhold, B., Prior, D., Suhayda, J., Keller, G., Yang, Z., and Fan, Y., 1988. Marine dispersal and deposition of Huanghe silts by gravity-driven underflows.,332 (6165): 629-632.

Wu, L., Wu, D., Lin, X., and Chao, M., 2009. The study of the Yellow Sea Warm Current and its seasonal variability., 21 (2): 159-165.

Wu, X., 2016. Evolution of the Yellow River (Huanghe) delta and its responses to human activities at different time scales: From Holocene to Anthropocene. PhD thesis.Ocean University of China.

Wu, W., and Ge, Q., 2005. The possibility of occurring of the extraordinary floods on the eve of establishment of the Xia dynasty and the historical truth of Dayu’s successful regulating of floodwaters., 25 (6): 741-749 (in Chinese with English abstract).

Xiao, S., Li, A., Liu, J., Chen, M., Xie, Q., Jiang, F., Li, T., Xiang, R., and Chen, Z., 2006. Coherence between solar activity and the East Asian Winter Monsoon variability in the past 8000 years from Yangtze River-derived mud in the East China Sea., 37 (2-4): 293-304.

Xiang, R., Yang, Z., Satio, Y., Guo, Z., Fan, D., Li, Y., Xiao, S., Shi, X., and Chen, M., 2006. East Asia Winter Monsoon changes inferred from environmentally sensitive grain-size component records during the last 2300 years in mud area southwest off Cheju Island, ECS., 49 (6): 604-614.

Xu, Z., Lim, D., Choi, J., Yang, S., and Jung, H., 2009. Rare earth elements in bottom sediments of major rivers around the Yellow Sea: Implications for sediment provenance., 29 (5): 291-300.

Xu, Z., Li, T., Chang, F., Choi, J., Lim, D., and Xu, F., 2012. Sediment provenance discrimination in northern Okinawa Trough during the last 24kyBP and paleo-environmental implication: Rare earth elements evidence., 30 (11): 1184-1190.

Xu, Z., Li, T., Chang, F., Wan, S., Choi, J., and Lim, D., 2014. Clay-sized sediment provenance change in the northern Okinawa Trough since 22kyrBP and its paleoenvironmental implication.,399: 236-245.

Xue, C., Zhou, Y., and Zhu, X., 2004. The Huanghe River course and delta from end of late Pleistocene to the 7th Century BC., 26 (1): 48-61 (in Chinese with English abstract).

Yang, S., and Youn, J., 2007. Geochemical compositions and provenance discrimination of the central South Yellow Sea sediments., 243 (1): 229-241.

Yang, S., Jung, H., Man, S., and Li, C., 2002. The rare earth ele- ment compositions of the Changjiang (Yangtze) and Huanghe (Yellow) River sediments.,201 (2): 407-419.

Yang, S., Li, C., and Lee, C., 2003. REE geochemistry of suspended sediments from the rivers around the Yellow Sea and provenance indicators., 48 (11): 1135- 1139.

Yang, S., Li, C., Jung, H., and Lee, H., 2002. Discrimination of geochemical compositions between the Changjiang and the Huanghe sediments and its application for the identification of sediment source in the Jiangsu coastal plain, China., 186 (3-4): 229-241.

Yang, Z., and Liu, J., 2007. Aunique Huanghe derived distal subaqueous delta in the Yellow Sea., 240 (1-4): 169-176.

Yuan, D., Zhu, J., Li, C., and Hu, D., 2008. Cross-shelf circulation in the Yellow and East China Seas indicated by MODIS satellite observations., 70 (1-2): 134- 149.

Yuan, Z., 1998.. Henan Renmin Press, Zhengzhou, 66-73.

Yuan, Z., Xiao, X., Wang, F., Xing, L., Wang, Z., Zhang, H., Xiang, D., Zhou, L., and Zhao, L., 2018. Spatiotemporal temperature variations in the East China Sea shelf during the Holocene in response to surface circulation evolution., 482: 46-55.

Zang, J., Tang, Y., Zou, E., and Lie, H., 2003. Analysis of Yellow Sea circulation., 48 (z1): 12-20.

Zhang, S., Wang, Q., Lu, Y., Cui, C., and Yuan, Y., 2008. Observation of the seasonal evolution of the Yellow Sea Cold Water Mass in 1996–1998., 28 (3): 442-457.

Zhang, Y., Huang, C., Pang, J., Zha, X., Zhou, Y., and Gu, H., 2013. Holocene paleofloods related to climatic events in the upper reaches of the Hanjiang River valley, middle Yangtze River Basin, China., 195: 1-12.

Zhao, M., Ding, L., Xing, L., Qiao, S., and Yang, Z., 2014. Major mid-late Holocene cooling in the East China Sea revealed by an Alkenone sea surface temperature record., 13 (6): 935-940.

Zhao, Y., Park, Y., Qin, Y., Choi, J., Gao, S., Li, F., Cheng, P., and Jiang, R., 2001. Material source for the eastern Yellow Sea mud: Evidence of mineralogy and geochemistry from China-Korea joint investigations., 7 (1): 22- 26.

Zhao, Y., Zou, X., Liu, Q., Wang, C., Ge, C., and Xu, M., 2018. Clay mineralogy indicates the muddy sediment provenance in the estuarine-inner shelf of the East China Sea., 152: 69-79.

Zhou, C., Dong, P., and Li, G., 2015. Hydrodynamic processes and their impacts on the mud deposit in the southern Yellow Sea., 360: 1-16.

Zou, Z., Liu, K., Zhao, W., Zhou, H., Meng, X., Li, Y., and Gurrola, H., 2016. Upper crustal structure beneath the northern South Yellow Sea revealed by wide-angle seismic tomography and joint interpretation of geophysical data., 51 (S1): 108-122.

. Tel: 0086-532-66782950

E-mail: hjwang@ouc.edu.cn

April29, 2019;

May 13, 2019;

May 27, 2019

(Edited by Chen Wenwen)